NSA Plans Five Mission Centers in Rapid Agency-Wide Reorganization

Directorate structure gives way to mission-focused centers

The U.S. National Security Agency is reorganizing around five mission centers covering China, cybersecurity, artificial intelligence, combat support and global intelligence.

The restructuring is intended to move intelligence and related capabilities to military customers and operational theaters more quickly. It will replace the NSA’s existing directorate structure, changing how personnel, missions and leadership are organized across the agency.

Army Gen. Joshua Rudd, who leads both the NSA and U.S. Cyber Command, briefed employees earlier this month and initiated a 30-day implementation period. The changes were first publicly reported on September 13, 2026.

The NSA expects the five centers to reach full operational capability by January. In this context, that means each center should be able to perform its assigned missions, rather than merely existing on an organizational chart.

Several consequential details remain unsettled, however. The final reporting lines, placement of existing units and effects on the agency’s resource-sharing relationship with Cyber Command have not been disclosed.

The NSA did not respond to a request for comment.

Five chiefs will oversee the new structure

Each mission center will have its own chief. Some leadership selections have reportedly already been made, including the person chosen to direct the artificial intelligence center. Those appointments could be communicated internally as soon as Monday.

The five planned centers are:

• China, focusing the agency’s intelligence capabilities on its most prominent nation-state priority.
• Cybersecurity, bringing together work associated with cyber defense and related intelligence missions.
• Artificial intelligence, giving AI a dedicated organizational position rather than treating it solely as a supporting technology.
• Combat support, intended to connect NSA capabilities more directly with military operations and operational theaters.
• Global intelligence, covering intelligence responsibilities that do not sit within the more narrowly defined centers.

Precise staffing levels, budgets and internal subdivisions have not been disclosed. It is also unknown whether personnel will be physically reassigned, placed under new administrative chains or reorganized mainly through changes in command authority.

Rudd has identified speed as his leading priority, followed by scale, innovation and integration. He described those priorities during his second Senate Intelligence Committee confirmation hearing in January and was confirmed to the dual NSA–Cyber Command role in March.

NSA Deputy Director Tim Kosiba and Executive Director Darren Turner reportedly share the objective of accelerating delivery to military users. The central test will be whether the new structure shortens decision and production cycles without creating additional approval layers.

Tailored Access Operations is expected to move

One of the most significant anticipated changes concerns Tailored Access Operations, or TAO, the NSA’s elite hacking unit. TAO is expected to become part of the global intelligence center.

Created in the 1990s and recently revived, TAO conducts highly specialized offensive access operations. Moving it into the global intelligence center would place a major operational capability inside the broadest of the five new organizations.

The exact command relationship has not been made public. It is unclear whether TAO would retain substantial autonomy, become fully integrated into the center’s management structure or share responsibilities with other offensive and intelligence components.

The future of the Cybersecurity Collaboration Center is also unresolved. That organization was established to improve collaboration with outside cybersecurity partners, but its position within the five-center model has not been disclosed.

Other offices created in recent years may face similar uncertainty. Until the agency publishes or internally finalizes a detailed structure, personnel may know their broad mission center without knowing their permanent reporting chain, support organization or operational authority.

That ambiguity matters at an intelligence agency where collection, analysis, cybersecurity and operational functions depend on controlled access to data, specialized infrastructure and clearly delegated authorities. A nominal transfer is straightforward. Rebuilding workflows and accountability is not.

The NSA–Cyber Command relationship adds complexity

The reorganization also intersects with the unusually close relationship between the NSA and U.S. Cyber Command. Rudd leads both organizations, while they share personnel, resources and a campus at Fort Meade, Maryland.

No final explanation has been provided for how the five centers will interact with Cyber Command. The unanswered questions include how shared personnel will be assigned, which organization will prioritize competing requirements and whether existing support arrangements will remain intact during the transition.

The new combat support center could improve the movement of intelligence into military operations. A dedicated cybersecurity center could likewise make certain defensive missions easier to coordinate. But organizational labels alone do not resolve questions of tasking, operational control or resource allocation.

The compressed schedule magnifies those concerns. Current and former NSA officials expect the realignment to disrupt established processes, and Rudd and Kosiba reportedly recognize that operational and bureaucratic problems will emerge during implementation.

Temporary friction could take several forms: delayed approvals, duplicated work, uncertainty over ownership of an intelligence requirement or gaps between units operating under old and new chains of command. No specific disruption has been publicly confirmed.

The risk is therefore organizational rather than a disclosed technical compromise. There is no associated vulnerability, patch, indicator of compromise or defensive control for external organizations to deploy.

Earlier overhauls offer a cautionary comparison

The NSA last attempted a restructuring on a comparable scale roughly a decade ago through an initiative known as NSA21. That effort sought to combine offensive and defensive organizations.

Current and former officials later characterized NSA21 as unsuccessful. Rather than simplifying operations, it was associated with greater bureaucratic complexity, employee frustration and confusion inside the agency.

A later plan called EVO25 began around the start of the second Trump administration. It would have consolidated NSA offices, but many officials viewed it primarily as an attempt to prevent the U.S. DOGE Service from imposing its own restructuring. EVO25 was eventually abandoned.

Those precedents do not determine the outcome of the five-center model. The current plan differs by aligning the agency around defined missions and operational priorities, including dedicated centers for China and AI.

They do expose a recurring implementation problem. Consolidating offices can reduce fragmentation, but it can also place dissimilar functions under a single management layer. Combining capabilities on paper may create new handoffs rather than eliminating old ones.

The 30-day implementation period leaves limited time to identify those effects before the structure is in place.

Success depends on unresolved command decisions

The reorganization’s stated benefits are speed, scale, innovation and integration. A mission-center structure could make intelligence production more responsive by placing analysts, technical specialists and operational capabilities under leaders responsible for specific outcomes.

Whether it achieves that goal will depend on decisions that remain undisclosed. The agency must determine where existing organizations belong, how center chiefs divide authority and how shared NSA–Cyber Command resources are prioritized.

Leadership appointments will provide an early indication of how the model is intended to work. The eventual placement of the Cybersecurity Collaboration Center and the final status of TAO will be more substantive signals.

For now, there are no technical mitigations for industry, government agencies or individual users to implement. The relevant safeguards are internal: phased execution, clearly designated leaders, preserved operational continuity and rapid correction of problems discovered during the transition.

The NSA plans to have all five centers operational by January. Meeting that target will be only the first measure of progress. The harder question is whether intelligence reaches military users faster without sacrificing coordination, clarity or capability along the way.